Cohere Transcribe Arabic is an open-source model built for Arabic's toughest transcription problems
Cohere has launched Transcribe Arabic, an open-source speech recognition model designed for Arabic dialects and bilingual speech, outperforming Whisper and OmniASR.
- Cohere released Transcribe Arabic, an open-source speech recognition model with 2 billion parameters.
- The model outperforms Whisper and OmniASR in Arabic dialect and bilingual speech transcription.
- Available on Hugging Face under the Apache 2.0 license for community use and development.
- Aims to address gaps in Arabic speech recognition where dialects and code-switching are challenging.
Cohere has introduced Transcribe Arabic, an open-source speech recognition model tailored for Arabic dialects and bilingual speech. The model, which boasts 2 billion parameters, is now available on Hugging Face under the Apache 2.0 license. According to Cohere, it outperforms existing models like Whisper and OmniASR in handling challenging Arabic transcription tasks, including code-switching between Arabic and English.
The release addresses a significant gap in Arabic speech recognition, where dialects and mixed-language speech have historically posed difficulties for automated systems. By making the model open-source, Cohere aims to foster broader adoption and further improvements through community contributions. The model's performance claims are based on internal benchmarks, though independent validation is still pending.
Source: Cohere Transcribe Arabic is an open-source model built for Arabic's toughest transcription problems. Read the full piece at the source.
Provides a high-performance, open-source tool for Arabic speech recognition, enabling new applications in NLP and AI.
Opens opportunities for companies targeting Arabic-speaking markets with advanced transcription services.
Offers a practical resource for learning about speech recognition and Arabic NLP.
Expands accessibility to advanced AI tools for Arabic language processing.
- code-switching
- The practice of alternating between two or more languages or dialects within a conversation or sentence.
AI ToolsI connected ChatGPT to my bank, and it's my go-to finance app now - here's how (and why)
AI ToolsThe AI Bill Grows in the Agent Loop
AI ToolsBuild a unified semantic layer across datasets with multi-dataset Topics in Amazon Quick
AI ToolsAnthropic's Claude Cowork AI agent is now available on mobile and web
AI ToolsBuild a serverless image editing agent with Amazon Bedrock AgentCore harness
BusinessCopilot goes cheap as Microsoft phases out OpenAI and Anthropic models to cut costs
Microsoft is swapping OpenAI and Anthropic models for its own MAI models in Copilot to reduce costs, potentially impacting performance for users.
Canada province preparing lawsuit against OpenAI over school shooting - Yahoo
A Canadian province is preparing a lawsuit against OpenAI, alleging the company's AI systems contributed to a school shooting incident.
Tyndall Airmen showcase AI innovations - acc.af.mil
Tyndall Air Force Base pilots demonstrated AI-driven innovations for military applications, including autonomous systems and predictive maintenance.
New artificial intelligence model reveals invisible multiple sclerosis lesions - News-Medical
Researchers developed an AI model that detects previously invisible multiple sclerosis lesions in MRI scans, potentially transforming early diagnosis.
Grants support new research on role of AI in the arts - Binghamton University
Binghamton University has awarded grants to fund new research exploring how AI intersects with artistic expression and cultural impact.
When AI hurts people, who’s to blame? Global experts grapple with accountability - UN News
The United Nations is convening global experts to address legal and ethical gaps in assigning blame when AI systems cause harm.